Papers › Global Self-Attention as a Replacement for Graph Convolution

Global Self-Attention as a Replacement for Graph Convolution

7 Aug 2021arXiv:2108.03348archive 2025-07-28

Md Shamim Hussain, Mohammed J. Zaki, Dharmashankar Subramanian

We propose an extension to the transformer neural network architecture for general-purpose graph learning by adding a dedicated pathway for pairwise structural information, called edge channels. The resultant framework - which we call Edge-augmented Graph Transformer (EGT) - can directly accept, process and output structural information of arbitrary form, which is important for effective learning on graph-structured data. Our model exclusively uses global self-attention as an aggregation mechanism rather than static localized convolutional aggregation. This allows for unconstrained long-range dynamic interactions between nodes. Moreover, the edge channels allow the structural information to evolve from layer to layer, and prediction tasks on edges/links can be performed directly from the output embeddings of these channels. We verify the performance of EGT in a wide range of graph-learning experiments on benchmark datasets, in which it outperforms Convolutional/Message-Passing Graph Neural Networks. EGT sets a new state-of-the-art for the quantum-chemical regression task on the OGB-LSC PCQM4Mv2 dataset containing 3.8 million molecular graphs. Our findings indicate that global self-attention based aggregation can serve as a flexible, adaptive and effective replacement of graph convolution for general-purpose graph learning. Therefore, convolutional local neighborhood aggregation is not an essential inductive bias.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2108.03348")

Code

Syntology Ran 0 of 4 code samples harvested from 1 repository linked to this paper; 4 have no recorded run.

By repository: official repository: 4 samples from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

shamim-hussain/egt officialmentioned in papermentioned on GitHubtfMIT report
shamim-hussain/egt_pytorch officialmentioned on GitHubpytorch report
shamim-hussain/egt_triangular officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

4 samples harvested; 0 ran; 0 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

4unverified

Licence: 0 of the 4 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from shamim-hussain/egt. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

import_object shamim-hussain/egt/lib/training/importer.py official repository unverified MIT (permissive) · 23012f87aaa39c42 · report
import_scheme shamim-hussain/egt/lib/training/importer.py official repository unverified MIT (permissive) · ca5300d8a50237fe · report
read_config_from_file shamim-hussain/egt/lib/training/training_base.py official repository unverified MIT (permissive) · 4b0f182cd45413d3 · report
save_config_to_file shamim-hussain/egt/lib/training/training_base.py official repository unverified MIT (permissive) · 219b23ce1676877d · report

Tasks

Edge ClassificationGraph ClassificationGraph LearningGraph Property PredictionGraph RegressionInductive BiasLink PredictionNode ClassificationTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification CIFAR10 100k EGT Accuracy (%) 68.702 #15 of 20 Archive leaderboard report
Graph Classification MNIST EGT Accuracy 98.173 #10 of 13 Archive leaderboard report
Graph Property Prediction ogbg-molhiv EGT Test ROC-AUC 0.806 ± 0.0065 #13 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molpcba EGT Test AP 0.2961 ± 0.0024 #16 of 36 Archive leaderboard report
Graph Regression PCQM4M-LSC EGT Validation MAE 0.1224 #10 of 11 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC EGT + Triangular Attention Test MAE 0.0683 #2 of 20 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC EGT + Triangular Attention Validation MAE 0.0671 #2 of 20 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC EGT Test MAE 0.0862 #10 of 20 Archive leaderboard report
Graph Regression PCQM4Mv2-LSC EGT Validation MAE 0.0857 #10 of 20 Archive leaderboard report
Graph Regression ZINC 100k EGT MAE 0.143 #4 of 8 Archive leaderboard report
Graph Regression ZINC-500k EGT MAE 0.108 #24 of 36 Archive leaderboard report
Link Prediction TSP/HCP Benchmark set EGT F1 0.853 #3 of 4 Archive leaderboard report
Node Classification CLUSTER EGT Accuracy 79.232 #2 of 12 Archive leaderboard report
Node Classification PATTERN EGT Accuracy 86.821 #5 of 11 Archive leaderboard report
Node Classification PATTERN 100k EGT Accuracy (%) 86.816 #1 of 9 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: EGT

Absolute Position EncodingsAdamAttentionBPEConvolutionDense ConnectionsDropoutEGTGraph Convolutional NetworksGraph TransformerLabel SmoothingLapEigenLaplacian PELayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections